One of the crucial ingredients for the construction of a subunit vaccine based on recombinant proteins or peptides is an adjuvant. However, today the only adjuvant available for human use is alum, which has not been effective in the animal IV-1 vaccine-trials attempted thus far. One of the most powerful adjuvants used, is the so called ISCOM adjuvant, which is made from a glycoside saponin preparation called Quil-A and cholesterol. This adjuvant recently demonstrated encouraging results in HIV-1 vaccine trials on chimpanzees, where it raised group specific responses against gp120 as well as giving an immune response stronger than those seen in animals infected with the virus. The Quil-A mixture is, however, to undefined for human use and many questions as to how this adjuvant excerpt its effect are unknown. We therefore initiated a search for an characterization of the individual molecules in the Quil-A mixture that were responsible for the adjuvant effect, in order to obtain a more defined preparation. An additional advantage of this strategy would be that potentially toxic non-active components could be eliminated. The long term objectives of this proposal would be that the structural features of the ISCOM particle itself, as well as its active components, can be studied which would greatly improve the possibility to optimize this adjuvant for use in various human vaccines. Our preliminary results indicate that only 2 or 3 components are necessary for forming functional ISCOM particles, of the at least 50 components present in the Quil-A mixture. Furthermore, the adjuvant property and the ISCOM particle forming property seem to reside in different molecular species, which would make it feasible to optimize the ISCOM adjuvant system for various antigens by varying the composition of the participating molecules. Our aim therefore is to develop a new, highly defined, subunit-vaccine adjuvant system for HIV-1 that is based on the well characterized, and for many viral antigens superiorly working, ISCOM system.